An image capturing apparatus includes an image pickup device configured to output image data, and at least one processor programmed to perform the operations of following units: a calculation unit configured to calculate an evaluation value used to determine whether to perform an image capturing operation for recording the image data; a setting unit configured to set a threshold value used to determine whether to perform an image capturing operation for recording the image data; a determination unit configured to make a determination as to whether to control execution of an image capturing operation using the evaluation value and the threshold value; and a storing unit configured to store image capturing history information obtained from execution of an image capturing operation based on the determination made by the determination unit, wherein the setting unit sets the threshold value based on the image capturing history information.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
3. The image capturing apparatus according to claim 2, wherein the information about a subject is at least one of information about a sound and information that is based on image data captured by the image pickup device.
4. The image capturing apparatus according to claim 1, wherein an initial value of the threshold value is set based on a result of past learning.
5. The image capturing apparatus according to claim 1, wherein the setting unit makes a comparison between time at which the latest image capturing operation was performed stored by the storing unit and current time, and, if a difference obtained by the comparison is smaller than a predetermined value, the setting unit sets the threshold value higher than an initial value thereof.
6. The image capturing apparatus according to claim 1, wherein the setting unit sets the threshold value to an initial value thereof in a case where current time is within a predetermined time from time of the latest image capturing operation stored by the storing unit and a change of the evaluation value stored by the storing unit has an increasing tendency.
7. The image capturing apparatus according to claim 1, wherein the setting unit sets the threshold value in such a manner that the threshold value decreases as image capturing time passes.
8. The image capturing apparatus according to claim 7, wherein, when decreasing the threshold value as image capturing time passes, the setting unit sets the threshold value to an initial value thereof in a case where an image capturing operation has been performed based on the determination made by the determination unit in a state in which the threshold value has become lower than the initial value.
9. The image capturing apparatus according to claim 1, wherein, for a predetermined time after image capturing, the setting unit sets the threshold value higher than an initial value thereof.
10. The image capturing apparatus according to claim 1, wherein the setting unit sets the threshold value depending on still image capturing or moving image capturing.
11. The image capturing apparatus according to claim 1, wherein the setting unit includes a discrimination unit configured to discriminate a state of the image capturing apparatus, and sets the threshold value depending on discrimination performed by the discrimination unit.
12. The image capturing apparatus according to claim 1, wherein the determination unit determines whether to perform an image capturing operation by a neural network, and causes learning to be performed by changing weights of the neural network based on learning information included in image data to be learned.
This invention relates to an image capturing apparatus that uses a neural network to determine whether to perform an image capturing operation. The apparatus includes a determination unit that evaluates input data to decide if an image should be captured. The neural network within the determination unit processes the input data and adjusts its weights based on learning information embedded in the image data to be learned. This adaptive learning mechanism allows the neural network to improve its decision-making over time by refining its parameters in response to the learning data. The apparatus may also include an image capturing unit that executes the capture operation when the determination unit approves it. The learning process involves modifying the neural network's weights to enhance accuracy in future determinations. This system is designed to optimize image capture decisions by leveraging machine learning, reducing unnecessary captures and improving efficiency in scenarios where selective imaging is required. The apparatus may be used in applications such as surveillance, autonomous systems, or any scenario where intelligent image capture is beneficial. The neural network's ability to learn from past data ensures continuous improvement in decision-making performance.
13. The image capturing apparatus according to claim 1, wherein the determination unit determines an image capturing method by a neural network, and causes learning to be performed by changing weights of the neural network based on learning information included in image data to be learned.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
September 25, 2020
November 1, 2022
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.